Creating Multiple DataFrames with Pandas

Question

Is there a way to use a 'for loop' or something similar to run through the code to create multiple DataFrames with Pandas that I can assign to separate variables, instead of hardcoding both DataFrames?

If I add new tickers it wouldn't be efficient to keep hardcoding them.

import pandas_datareader as pdr
from datetime import datetime

Equity_Tickers = ["FB", "MSFT"]

start = datetime(2018, 9, 15)
end = datetime.today().date()

# First DataFrame 
data = pdr.DataReader(Equity_Tickers[0], 'yahoo', start, end)
df = data[['Adj Close']]

# Second DataFrame
data = pdr.DataReader(Equity_Tickers[1], 'yahoo', start, end)
df1 = data[['Adj Close']]

Answer 1

You could certainly use a variable instead of hard-coding the ticker:

ticker = Equity_Tickers[0]  # FB
data = pdr.DataReader(ticker, 'yahoo', start, end)

To create a loop, you could do something like:

for ticker in Equity_Tickers:
    data = pdr.DataReader(ticker, 'yahoo', start, end)
    df = data[['Adj Close']]

The tricky thing is storing the df variable in each iteration of the loop instead of overwriting it. One solution is to store each DataFrame in a list or dictionary.

frames = []
for ticker in Equity_Tickers:
    data = pdr.DataReader(ticker, 'yahoo', start, end)
    frames.append(data[['Adj Close']])

# eg, use frames[0] to access first ticker's DataFrame

OR

frames = {}
for ticker in Equity_Tickers:
    data = pdr.DataReader(ticker, 'yahoo', start, end)
    frames[ticker] = data[['Adj Close']]

# eg, use frames['FB'] to access FB ticker's DataFrame

If you're feeling really fancy you could also use a list comprehension / dictionary comprehension to accomplish this in one line:

# list comprehension
frames = [pdr.DataReader(t, 'yahoo', start, end)[['Adj Close']] for t in Equity_Tickers]

# dict comprehension
frames = {t: pdr.DataReader(t, 'yahoo', start, end)[['Adj Close']] for t in Equity_Tickers}

Additional note: You might also consider combining the frames into one large DataFrame, perhaps a tidy-style DataFrame where "ticker" is a field for each row.